The utility of portable nerve conduction testing for patients with carpal tunnel syndrome: A prospective clinical study The usefulness
and accuracy of a portable instrument,
the electroneurometer,
for measuring
distal motor latencies was determined in a prospective study of 28 patients (51 hands) with carpal tunnel syndrome and 10 controls (18 hands). There was a close correlation of distal motor latencies from the electroneurometer
with those from formal electrodiagnostic
testing. The average distal
motor latency of controls was 3.3 + 0.4 msec, compared with a mean value of 5.2 t in symptomatic
when combined with quantitative the neurometer
1.8 msec
bands. The sensitivity of the neurometer test alone was 69% (9 false-negatives); sensibility testing, sensitivity increased to 84%. Specificity of
test alone was 100% (no false-positives).
The portable electroneurometer
is a
device for screening patients with carpal tunnel syndrome. Formal electrodiagnostic testing is indicated for differentiation of lesions present at different levels, for detection of subtle sensory changes, and for use in those patients in whom surface electrode usage is ineffective. (J HAND SURG 1992;17A:77-81.) convenient,
painless,
inexpensive
David R. Steinberg, MD, Sacramento, C&j-. , Richard H. Gelberman, Boston, Muss., Bjorn Rydevik, MD, PhD, Giiteborg, Sweden, and Goran Lundborg, Mulmii, Sweden
A
lthough characteristic history and physical examination findings remain the cornerstones of diagnosis, sophisticated sensibility tests, provocative maneuvers, and electrodiagnostic studies have been advocated to provide quantitative documentation of median nerve compression at the wrist. Formal electrodiagnostic studies have been shown to provide the earliest objective evidence of carpal tunnel syndrome. ‘. ’ However, they are expensive, painful, and timeconsuming for the patient. The Nervepace electroneurometer (Neurotron Medical, Lawrenceville, N.J.), a hand-held device designed for use in the office, has
From the Department of Orthopaedic Hospital, Boston, Mass. Received for publication 4. 1991.
Surgery, Massachusetts
General
Jan. 18, 1991; accepted in revised form June
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Richard H. Gelberman, MD, Department of Orthopaedic Surgery, Massachusetts General Hospital, WACC - 527. Boston, MA 02114. 3/l/31515
MD,
been advocated as a simple, cost-effective means of measuring distal motor latency of the median nerve. The purpose of this study is to evaluate the usefulness and accuracy of the Nervepace electroneurometer as compared with sensibility testing and formal electrodiagnostic studies. Materials
and methods
Twenty-eight consecutive patients (46 symptomatic hands) whose history and examination findings were consistent with carpal tunnel syndrome were studied prospectively. There were 17 women and 11 men, ranging in age from 25 to 76 years (mean, 50 years). Both hands were involved in 18 patients; one hand was involved in the remaining 10. All patients had paresthesia, and there was numbness in all digits or in the median nerve distribution. Thirteen had nocturnal symptoms. Duration of symptoms ranged from 2 weeks to 10 years (mean, 28 months). Most patients noted easy fatigability with fine motor activities, such as writing or buttoning clothing. Thirty-six percent of the subjects had coexisting cervical arthritis, diabetes mellitus, or hypothyroidism. In 18 patients there was a positive Tinel’s sign over the median nerve at the wrist, and in 22 a wrist flexion test (Phalen’s test) was positive at
THE JOURNAL OF HAND SURGERY
77
78
The Journal of HAND SURGERY
Steinberg et al.
12
l
Formal Conduction Studies Portable Conduction Studies
. l
8
6 4
2
0
I
I
10
I
I
I
20 Nerve
I
30 Studies
I
I
40 (number)
I
I
50
I
I
60
Fig. 1.
60 seconds. Five patients had thenar weakness and atrophy. Seven patients had two-point discrimination greater than 6 mm. Twenty patients underwent Semmes-Weinstein monofilament testing by an occupational therapist with the wrist at 0 degrees of flexion and after 1 minute of gravity-assisted palmar flexion as described by Koris et al.3 All patients underwent formal electrodiagnostic testing of the median nerve in the Neurology Department. This included the asymptomatic extremities in five patients with unilateral carpal tunnel syndrome. Abnormal values consisted of distal motor latencies of 4.5 msec or greater and distal sensory latencies of 3.5 msec or greater.4 Ten healthy volunteers ( 18 hands) without symptoms of carpal tunnel syndrome served as controls. There were five women and five men, with ages ranging from 31 to 56 years (mean, 43 years). Formal electrodiagnostic testing was not performed on these subjects. The median nerves of all patients and controls were studied with the Nervepace electroneurometer, as described by Rosier and Blair (21st annual Rocky Moun-
tain Bioengineering Symposium and International ISA Biomedical Sciences Instrumentation Symposium, Boulder, Cola., 1984). Terminal motor latencies over the carpal tunnel were determined by stimulation of the median nerve 3 cm proximal to the distal flexion crease of the wrist. The recording silver/silver chloride electrocardiogram-type electrode with contact gel was placed over the abductor pollicis brevis. The reference electrode was placed on the palmar radial aspect of the proximal phalanx of the thumb, and the ground electrode was placed on the dorsum of the hand. The stimulus pulse was delivered to the skin through two stainless steel electrodes, 2 cm apart, placed longitudinally over the nerve. The skin was cleansed with alcohol before electrode application, and electrode gel (Spectra 360 electrode gel, Parker Laboratories, Orange, N.J.) was used to minimize skin resistance. The electroneurometer consists of a nerve stimulator producing a variable output of 0 to 300 volts of 0.5 msec duration. The stimulus pulse also triggers a counter to begin counting pulses from a 10 kHz square wave generator. The stimulus is conducted distally, causing muscle depolariza-
Vol. 17A, No. 1 January 1992
Portable nerve conduction testing
79
6
Formal Electrodiagnostic
Studies
Fig. 2. Correlatidn of distal motor latencies (msec) obtained by formal and portable electrodiagnostic testing. Each data point represents an individual hand. The solid line indicates the best-fit linear regression; r is the coefficient of correlation. (A value of 1.O indicates perfect correlation.) tion action potentials and a visible twitch of the abductor pollicis brevis. The counter is disabled when the recording electrodes pick up an amplified action potential greater than 10 mV, which occurs near, but not at, the onset of the action potential in the muscle. The number appearing on the counter is the time, in milliseconds, between the application of the stimulus and the depolarization of the muscle (i.e., the distal motor latency). During testing, stimulus intensity was sequentially increased until a vigorous muscle contraction was produced and the counter was disabled, representing latency of the supramaximal stimulus. The results of five consecutive readings were averaged to determine mean distal motor latency. Formal electrodiagnostic studies and electroneurometer testing were performed within an average time span of 4 weeks. Individual body temperatures were not measured. For control of environmental conditions all portable nerve conduction tests were performed in the same room, maintained at 72” F.
Results The mean distal motor latency measured by the neurometer in the 18 control hands was 3.3 ~fr 0.4 msec.
In the five asymptomatic hands, which underwent electrical testing both formally in the laboratory and in the office, distal motor latency was 3.7 + 0.2 msec, compared with neurometer measurements of 3.5 + 0.2 msec. The difference in neurometer latencies between the 18 controls and the 5 asymptomatic hands was not statistically significant. Fifty-one extremities (46 symptomatic and 5 asymptomatic) underwent formal electrodiagnostic studies and nerve conduction testing in the office. The mean difference between distal motor latencies for matched pairs of measurements was 0.3 msec with a standard deviation of 0.8 msec (range, - 1.1 to + 2.5 msec, Fig. 1). The results of these two tests were closely correlated, with a correlation coefficient of 0.93 (Fig. 2). This indicates that the electroneurometer accurately measures median nerve distal motor latency when compared with formal electrodiagnostic testing (a correlation coefficient of 1.00 indicates perfect positive correlation). In 46 symptomatic hands, mean distal motor latency determined by formal testing was 5.1 + 2.0 msec, compared with office values of 5.2 + 1.8 msec. These values were significantly different from values for con-
80
Steinberg et al.
trol hands (p = 0.0001). In 24 of these symptomatic hands, which also had abnormal distal motor latencies on formal testing, mean values were 6.4 & 1.8 msec, compared with mean neurometer readings of 6.5 + 1.5 msec. These values were significantly different from controls (p = 0.00005). Thirty-five symptomatic hands also underwent Semmes-Weinstein monofilament testing. With the wrist in neutral, all but one had values >2.83 (mean value, 3.56). With gravity-assisted wrist flexion for 60 seconds, 24 of 35 symptomatic hands tested demonstrated an increase of at least one value, corresponding to a positive quantitative Phalen’s test as described by Koris et al3 Of these patients, four had normal electrodiagnostic studies and two had a combination of normal distal motor latencies and increased distal sensory latencies . With a patient with carpal tunnel syndrome defined as one who had positive history and physical examination findings, there were no false-positive readings. There were nine false-negative readings (values C4.5 msec). Therefore, the sensitivity of the neurometer test alone in this series was 69%, compared with 78% sensitivity for formal electrodiagnostic testing. When the neurometer results were combined with the results of the quantitative Phalen’s test, sensitivity increased to 84%. Inasmuch as there were no false-positive readings in this study, specificity was 100%. Surface recordings could not be obtained in three patients. One patient, a 71-year-old black woman with hypertension and insulin-dependent diabetes mellitus, had peripheral neuropathy diagnosed on the basis of formal electrodiagnostic criteria. Formal distal motor latencies were 4.6 and 6.5 msec. The second patient, a 46-year-old white man with severe tophaceous gout, had significant skin changes. The distal motor latency with formal studies on the symptomatic side was 5.7 msec; sensory readings were absent. The third patient, a 65-year-old right-handed woman, had sustained a distal radius fracture 6 months earlier and had had a median nerve neurapraxia. Although surface readings were obtained on the unaffected side, none could be obtained on the contralateral side. Discussion Previous reports on carpal tunnel syndrome have relied on abnormalities of electrodiagnostic studies as the diagnostic standard against which other tests and treatment results are compared. ‘* 4-6However, these tests are expensive ($750 at the Massachusetts General Hospital), inconvenient, and painful, and they may result in significant delays before the institution of appropriate
The Journal of HAND SURGERY
therapy. Many physicians therefore rely on clinical examination for the diagnosis of carpal tunnel syndrome.‘. ’ This study demonstrates a close correlation of motor latencies from the electroneurometer and those from formal electrodiagnostic testing. The 3.3 msec control value established in this study correlates more closely with the accepted normal values from electrodiagnostic studies in our institution than do data from Rosier and Blair’s study. However, controls in the latter study consisted of patients with clinical symptoms of ulnar or median neuropathy who were determined to be normal on the basis of values obtained by formal electrodiagnostic studies. Distal motor latency values from the electroneurometer were slightly higher than those obtained by formal electrodiagnostic testing. This was expected, as the Nervepace electroneurometer requires a slight increase in muscle action potential to stop the counter. With formal studies, actual potential onset is determined on the oscilloscope. A mean difference between matched pairs of 0.25 msec is consistent with that reported in other studies. An office-based test that is sensitive and specific for carpal tunnel syndrome is helpful in the diagnosis and management of this condition. It provides a convenient method of quantitatively following the nerve’s response to various treatment regimens through multiple successive studies. Previous studies have considered a decrease in sensory amplitude the most sensitive indicator of median nerve compression,1.9 followed by abnormal Serrmres-Weinstein monofilament test results and elevated vibratory laboratory thresholds.” 2.‘OKoris et al3 further improved the sensitivity and specificity of objective testing of carpal tunnel syndrome by combining the Semrnes-Weinstein monofilament test with the wrist-in-flexion (Phalen’s) test. Our study demonstrates the usefulness of a test that evaluates the motor component of median nerve function. The sensitivity of this test was 69%, and the specificity was 100%. In conjunction with the quantitative Phalen’s test, the sensitivity increased to 84%. Addition of the Semmes-Weinstein monofilament test raises the sensitivity of clinical testing to 91%.” There continue to be specific indications for formal electrodiagnostic testing. These studies are particularly helpful in differentiating lesions that are present at different levels (e.g., median nerve compression proximal to the carpal tunnel). Also, they may detect subtle sensory changes in patients with suggestive clinical findings and equivocal objective test values. For patients with severe peripheral neuropathies and those with skin changes that preclude the use of skin electrodes, needle
Vol. 17A, No. 1 January 1992
are available with laboratory equipment. Persons who could not be studied with the portable device represented only 10% to 15% of all patients seen at Massachusetts General Hospital with symptoms suggestive of carpal tunnel syndrome. Inasmuch as electrodiagnostic study results may be normal in 8% of patients with carpal tunnel syndrome” and motor values may be normal in up to 20% of these patients, I2 it is important not to rely solely on electrical testing values. For those patients in whom quantitative data are sought, however, Semmes-Weinstein monofilament testing (with the wrist in neutral and flexed positions) combined with Nervepace electroneurometer studies provides high levels of sensitivity and specificity. In addition, these studies help quantitate the severity of nerve compression and provide objective measurements of both sensory and motor changes, allowing the physician to modify treatment accordingly. The electroneurometer is convenient, relatively painless, and inexpensive. These factors make it an effective instrument for screening patients in the office and for serial examinations to follow the progress of patients treated by both conservative and operative methods. ”
Portable nerve conduction testing
or ring electrodes
4.
5.
6.
7.
8. 9.
10.
11.
REFERENCES Gelberman RH, Szabo RM, Williamson RV, Dimick MP. Sensibility testing in peripheral nerve compression syndromes. J Bone Joint Surg 1983;65A:632-8. Szabo RM, Gelberman RH, Dimick MP. Sensibility testing in patients with carpal tunnel syndrome. J Bone Joint Surg 1984;66A:60-4. Koris M, Gelberman RH, Duncan K, Boublik M, Smith B. Carpal Tunnel syndrome - Evaluation of a quantitative
12.
81
provocational diagnostic test. Clin Orthop 1990;251: 157-61. Thomas JE, Lambert EH, Cseuz KA. Electrodiagnostic aspects of carpal tunnel syndrome. Arch Neur 1967:63541. Harris CM, Tanner F, Goldstein MN, Pettee DS. The surgical treatment of the carpal tunnel syndrome correlated with preoperative nerve conduction studies. 3 Bone Joint Surg 1979;61A:93-8. Gelberman RH, Aronson D, Weisman MH. Carpal tunnel syndrome - Results of a prospective trial of steroid injection and splinting. J Bone Joint Surg 1980;62A: I 18 I84. Phalen GS Carpal tunnel syndrome - 17 years experience in diagnosis and treatment of 654 hands. J Bone Joint Surg 1966:48A:21 l-28. Phalen GS. The carpal tunnel syndrome - clinical evaluation of 598 hands. Clin Orthop 1972;83:29-40. Melvin JL, Schuchmann JA, Lanese RR. Diagnostic specificity of motor and sensory nerve conduction variables in the carpal tunnel syndrome. Arch Phys Med Rehabil 1973;54:69-74. Gellman H, Gelberman RH, Tan A, Bone MJ. Carpal tunnel syndrome: an evaluation of the provocative diagnostic tests. J Bone Joint Surg 1966;68A:735-7. Grundberg AB. Carpal tunnel decompression in spite of normal electromyography. J HAND SURG 1983:8: 348-9. Leblhuber F, Reisecker F, Witzmann A. Carpal tunnel syndrome: neurographical parameters in different stages of median nerve compression. Acta Neurochir (Wien) 1986;81:125-7.
13. Feirstein MS. The performance and usefulness of nerve conduction studies in the orthopaedic office. Orthop Clin North Am 1988:19:859-66.
and accuracy of a portable instrument,
the electroneurometer,
for measuring
distal motor latencies was determined in a prospective study of 28 patients (51 hands) with carpal tunnel syndrome and 10 controls (18 hands). There was a close correlation of distal motor latencies from the electroneurometer
with those from formal electrodiagnostic
testing. The average distal
motor latency of controls was 3.3 + 0.4 msec, compared with a mean value of 5.2 t in symptomatic
when combined with quantitative the neurometer
1.8 msec
bands. The sensitivity of the neurometer test alone was 69% (9 false-negatives); sensibility testing, sensitivity increased to 84%. Specificity of
test alone was 100% (no false-positives).
The portable electroneurometer
is a
device for screening patients with carpal tunnel syndrome. Formal electrodiagnostic testing is indicated for differentiation of lesions present at different levels, for detection of subtle sensory changes, and for use in those patients in whom surface electrode usage is ineffective. (J HAND SURG 1992;17A:77-81.) convenient,
painless,
inexpensive
David R. Steinberg, MD, Sacramento, C&j-. , Richard H. Gelberman, Boston, Muss., Bjorn Rydevik, MD, PhD, Giiteborg, Sweden, and Goran Lundborg, Mulmii, Sweden
A
lthough characteristic history and physical examination findings remain the cornerstones of diagnosis, sophisticated sensibility tests, provocative maneuvers, and electrodiagnostic studies have been advocated to provide quantitative documentation of median nerve compression at the wrist. Formal electrodiagnostic studies have been shown to provide the earliest objective evidence of carpal tunnel syndrome. ‘. ’ However, they are expensive, painful, and timeconsuming for the patient. The Nervepace electroneurometer (Neurotron Medical, Lawrenceville, N.J.), a hand-held device designed for use in the office, has
From the Department of Orthopaedic Hospital, Boston, Mass. Received for publication 4. 1991.
Surgery, Massachusetts
General
Jan. 18, 1991; accepted in revised form June
No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. Reprint requests: Richard H. Gelberman, MD, Department of Orthopaedic Surgery, Massachusetts General Hospital, WACC - 527. Boston, MA 02114. 3/l/31515
MD,
been advocated as a simple, cost-effective means of measuring distal motor latency of the median nerve. The purpose of this study is to evaluate the usefulness and accuracy of the Nervepace electroneurometer as compared with sensibility testing and formal electrodiagnostic studies. Materials
and methods
Twenty-eight consecutive patients (46 symptomatic hands) whose history and examination findings were consistent with carpal tunnel syndrome were studied prospectively. There were 17 women and 11 men, ranging in age from 25 to 76 years (mean, 50 years). Both hands were involved in 18 patients; one hand was involved in the remaining 10. All patients had paresthesia, and there was numbness in all digits or in the median nerve distribution. Thirteen had nocturnal symptoms. Duration of symptoms ranged from 2 weeks to 10 years (mean, 28 months). Most patients noted easy fatigability with fine motor activities, such as writing or buttoning clothing. Thirty-six percent of the subjects had coexisting cervical arthritis, diabetes mellitus, or hypothyroidism. In 18 patients there was a positive Tinel’s sign over the median nerve at the wrist, and in 22 a wrist flexion test (Phalen’s test) was positive at
THE JOURNAL OF HAND SURGERY
77
78
The Journal of HAND SURGERY
Steinberg et al.
12
l
Formal Conduction Studies Portable Conduction Studies
. l
8
6 4
2
0
I
I
10
I
I
I
20 Nerve
I
30 Studies
I
I
40 (number)
I
I
50
I
I
60
Fig. 1.
60 seconds. Five patients had thenar weakness and atrophy. Seven patients had two-point discrimination greater than 6 mm. Twenty patients underwent Semmes-Weinstein monofilament testing by an occupational therapist with the wrist at 0 degrees of flexion and after 1 minute of gravity-assisted palmar flexion as described by Koris et al.3 All patients underwent formal electrodiagnostic testing of the median nerve in the Neurology Department. This included the asymptomatic extremities in five patients with unilateral carpal tunnel syndrome. Abnormal values consisted of distal motor latencies of 4.5 msec or greater and distal sensory latencies of 3.5 msec or greater.4 Ten healthy volunteers ( 18 hands) without symptoms of carpal tunnel syndrome served as controls. There were five women and five men, with ages ranging from 31 to 56 years (mean, 43 years). Formal electrodiagnostic testing was not performed on these subjects. The median nerves of all patients and controls were studied with the Nervepace electroneurometer, as described by Rosier and Blair (21st annual Rocky Moun-
tain Bioengineering Symposium and International ISA Biomedical Sciences Instrumentation Symposium, Boulder, Cola., 1984). Terminal motor latencies over the carpal tunnel were determined by stimulation of the median nerve 3 cm proximal to the distal flexion crease of the wrist. The recording silver/silver chloride electrocardiogram-type electrode with contact gel was placed over the abductor pollicis brevis. The reference electrode was placed on the palmar radial aspect of the proximal phalanx of the thumb, and the ground electrode was placed on the dorsum of the hand. The stimulus pulse was delivered to the skin through two stainless steel electrodes, 2 cm apart, placed longitudinally over the nerve. The skin was cleansed with alcohol before electrode application, and electrode gel (Spectra 360 electrode gel, Parker Laboratories, Orange, N.J.) was used to minimize skin resistance. The electroneurometer consists of a nerve stimulator producing a variable output of 0 to 300 volts of 0.5 msec duration. The stimulus pulse also triggers a counter to begin counting pulses from a 10 kHz square wave generator. The stimulus is conducted distally, causing muscle depolariza-
Vol. 17A, No. 1 January 1992
Portable nerve conduction testing
79
6
Formal Electrodiagnostic
Studies
Fig. 2. Correlatidn of distal motor latencies (msec) obtained by formal and portable electrodiagnostic testing. Each data point represents an individual hand. The solid line indicates the best-fit linear regression; r is the coefficient of correlation. (A value of 1.O indicates perfect correlation.) tion action potentials and a visible twitch of the abductor pollicis brevis. The counter is disabled when the recording electrodes pick up an amplified action potential greater than 10 mV, which occurs near, but not at, the onset of the action potential in the muscle. The number appearing on the counter is the time, in milliseconds, between the application of the stimulus and the depolarization of the muscle (i.e., the distal motor latency). During testing, stimulus intensity was sequentially increased until a vigorous muscle contraction was produced and the counter was disabled, representing latency of the supramaximal stimulus. The results of five consecutive readings were averaged to determine mean distal motor latency. Formal electrodiagnostic studies and electroneurometer testing were performed within an average time span of 4 weeks. Individual body temperatures were not measured. For control of environmental conditions all portable nerve conduction tests were performed in the same room, maintained at 72” F.
Results The mean distal motor latency measured by the neurometer in the 18 control hands was 3.3 ~fr 0.4 msec.
In the five asymptomatic hands, which underwent electrical testing both formally in the laboratory and in the office, distal motor latency was 3.7 + 0.2 msec, compared with neurometer measurements of 3.5 + 0.2 msec. The difference in neurometer latencies between the 18 controls and the 5 asymptomatic hands was not statistically significant. Fifty-one extremities (46 symptomatic and 5 asymptomatic) underwent formal electrodiagnostic studies and nerve conduction testing in the office. The mean difference between distal motor latencies for matched pairs of measurements was 0.3 msec with a standard deviation of 0.8 msec (range, - 1.1 to + 2.5 msec, Fig. 1). The results of these two tests were closely correlated, with a correlation coefficient of 0.93 (Fig. 2). This indicates that the electroneurometer accurately measures median nerve distal motor latency when compared with formal electrodiagnostic testing (a correlation coefficient of 1.00 indicates perfect positive correlation). In 46 symptomatic hands, mean distal motor latency determined by formal testing was 5.1 + 2.0 msec, compared with office values of 5.2 + 1.8 msec. These values were significantly different from values for con-
80
Steinberg et al.
trol hands (p = 0.0001). In 24 of these symptomatic hands, which also had abnormal distal motor latencies on formal testing, mean values were 6.4 & 1.8 msec, compared with mean neurometer readings of 6.5 + 1.5 msec. These values were significantly different from controls (p = 0.00005). Thirty-five symptomatic hands also underwent Semmes-Weinstein monofilament testing. With the wrist in neutral, all but one had values >2.83 (mean value, 3.56). With gravity-assisted wrist flexion for 60 seconds, 24 of 35 symptomatic hands tested demonstrated an increase of at least one value, corresponding to a positive quantitative Phalen’s test as described by Koris et al3 Of these patients, four had normal electrodiagnostic studies and two had a combination of normal distal motor latencies and increased distal sensory latencies . With a patient with carpal tunnel syndrome defined as one who had positive history and physical examination findings, there were no false-positive readings. There were nine false-negative readings (values C4.5 msec). Therefore, the sensitivity of the neurometer test alone in this series was 69%, compared with 78% sensitivity for formal electrodiagnostic testing. When the neurometer results were combined with the results of the quantitative Phalen’s test, sensitivity increased to 84%. Inasmuch as there were no false-positive readings in this study, specificity was 100%. Surface recordings could not be obtained in three patients. One patient, a 71-year-old black woman with hypertension and insulin-dependent diabetes mellitus, had peripheral neuropathy diagnosed on the basis of formal electrodiagnostic criteria. Formal distal motor latencies were 4.6 and 6.5 msec. The second patient, a 46-year-old white man with severe tophaceous gout, had significant skin changes. The distal motor latency with formal studies on the symptomatic side was 5.7 msec; sensory readings were absent. The third patient, a 65-year-old right-handed woman, had sustained a distal radius fracture 6 months earlier and had had a median nerve neurapraxia. Although surface readings were obtained on the unaffected side, none could be obtained on the contralateral side. Discussion Previous reports on carpal tunnel syndrome have relied on abnormalities of electrodiagnostic studies as the diagnostic standard against which other tests and treatment results are compared. ‘* 4-6However, these tests are expensive ($750 at the Massachusetts General Hospital), inconvenient, and painful, and they may result in significant delays before the institution of appropriate
The Journal of HAND SURGERY
therapy. Many physicians therefore rely on clinical examination for the diagnosis of carpal tunnel syndrome.‘. ’ This study demonstrates a close correlation of motor latencies from the electroneurometer and those from formal electrodiagnostic testing. The 3.3 msec control value established in this study correlates more closely with the accepted normal values from electrodiagnostic studies in our institution than do data from Rosier and Blair’s study. However, controls in the latter study consisted of patients with clinical symptoms of ulnar or median neuropathy who were determined to be normal on the basis of values obtained by formal electrodiagnostic studies. Distal motor latency values from the electroneurometer were slightly higher than those obtained by formal electrodiagnostic testing. This was expected, as the Nervepace electroneurometer requires a slight increase in muscle action potential to stop the counter. With formal studies, actual potential onset is determined on the oscilloscope. A mean difference between matched pairs of 0.25 msec is consistent with that reported in other studies. An office-based test that is sensitive and specific for carpal tunnel syndrome is helpful in the diagnosis and management of this condition. It provides a convenient method of quantitatively following the nerve’s response to various treatment regimens through multiple successive studies. Previous studies have considered a decrease in sensory amplitude the most sensitive indicator of median nerve compression,1.9 followed by abnormal Serrmres-Weinstein monofilament test results and elevated vibratory laboratory thresholds.” 2.‘OKoris et al3 further improved the sensitivity and specificity of objective testing of carpal tunnel syndrome by combining the Semrnes-Weinstein monofilament test with the wrist-in-flexion (Phalen’s) test. Our study demonstrates the usefulness of a test that evaluates the motor component of median nerve function. The sensitivity of this test was 69%, and the specificity was 100%. In conjunction with the quantitative Phalen’s test, the sensitivity increased to 84%. Addition of the Semmes-Weinstein monofilament test raises the sensitivity of clinical testing to 91%.” There continue to be specific indications for formal electrodiagnostic testing. These studies are particularly helpful in differentiating lesions that are present at different levels (e.g., median nerve compression proximal to the carpal tunnel). Also, they may detect subtle sensory changes in patients with suggestive clinical findings and equivocal objective test values. For patients with severe peripheral neuropathies and those with skin changes that preclude the use of skin electrodes, needle
Vol. 17A, No. 1 January 1992
are available with laboratory equipment. Persons who could not be studied with the portable device represented only 10% to 15% of all patients seen at Massachusetts General Hospital with symptoms suggestive of carpal tunnel syndrome. Inasmuch as electrodiagnostic study results may be normal in 8% of patients with carpal tunnel syndrome” and motor values may be normal in up to 20% of these patients, I2 it is important not to rely solely on electrical testing values. For those patients in whom quantitative data are sought, however, Semmes-Weinstein monofilament testing (with the wrist in neutral and flexed positions) combined with Nervepace electroneurometer studies provides high levels of sensitivity and specificity. In addition, these studies help quantitate the severity of nerve compression and provide objective measurements of both sensory and motor changes, allowing the physician to modify treatment accordingly. The electroneurometer is convenient, relatively painless, and inexpensive. These factors make it an effective instrument for screening patients in the office and for serial examinations to follow the progress of patients treated by both conservative and operative methods. ”
Portable nerve conduction testing
or ring electrodes
4.
5.
6.
7.
8. 9.
10.
11.
REFERENCES Gelberman RH, Szabo RM, Williamson RV, Dimick MP. Sensibility testing in peripheral nerve compression syndromes. J Bone Joint Surg 1983;65A:632-8. Szabo RM, Gelberman RH, Dimick MP. Sensibility testing in patients with carpal tunnel syndrome. J Bone Joint Surg 1984;66A:60-4. Koris M, Gelberman RH, Duncan K, Boublik M, Smith B. Carpal Tunnel syndrome - Evaluation of a quantitative
12.
81
provocational diagnostic test. Clin Orthop 1990;251: 157-61. Thomas JE, Lambert EH, Cseuz KA. Electrodiagnostic aspects of carpal tunnel syndrome. Arch Neur 1967:63541. Harris CM, Tanner F, Goldstein MN, Pettee DS. The surgical treatment of the carpal tunnel syndrome correlated with preoperative nerve conduction studies. 3 Bone Joint Surg 1979;61A:93-8. Gelberman RH, Aronson D, Weisman MH. Carpal tunnel syndrome - Results of a prospective trial of steroid injection and splinting. J Bone Joint Surg 1980;62A: I 18 I84. Phalen GS Carpal tunnel syndrome - 17 years experience in diagnosis and treatment of 654 hands. J Bone Joint Surg 1966:48A:21 l-28. Phalen GS. The carpal tunnel syndrome - clinical evaluation of 598 hands. Clin Orthop 1972;83:29-40. Melvin JL, Schuchmann JA, Lanese RR. Diagnostic specificity of motor and sensory nerve conduction variables in the carpal tunnel syndrome. Arch Phys Med Rehabil 1973;54:69-74. Gellman H, Gelberman RH, Tan A, Bone MJ. Carpal tunnel syndrome: an evaluation of the provocative diagnostic tests. J Bone Joint Surg 1966;68A:735-7. Grundberg AB. Carpal tunnel decompression in spite of normal electromyography. J HAND SURG 1983:8: 348-9. Leblhuber F, Reisecker F, Witzmann A. Carpal tunnel syndrome: neurographical parameters in different stages of median nerve compression. Acta Neurochir (Wien) 1986;81:125-7.
13. Feirstein MS. The performance and usefulness of nerve conduction studies in the orthopaedic office. Orthop Clin North Am 1988:19:859-66.
